NEHA December 2022 Journal of Environmental Health

December 2022 • Journal of Environmental Health 13 isms that might be present when indicator bacteria are detected at concentrations above the water quality criteria. While the results reported here support past findings by CT DEEP that all of these watersheds have had fecal contamination (State of Connecticut, 2022), the lack of correlation between E. coli levels and GenBac3 counts presents challenges in identifying sources of E. coli. As the elevated E. coli levels were predominantly in the summer, one possibility is that the E. coli originated from avian sources, particularly geese, which are known to have gut microbiota fluctuations resulting in elevated E. coli levels in summer months (Alderisio & DeLuca, 1999). Additionally, birds are known to have low levels of Bacteroidetes, further supporting the hypothesis that geese or other birds could have contributed E. coli while not adding to the levels of Bacteroidetes detected. Another possible explanation for the lack of correlation is that conditions were more favorable to support E. coli persistence outside of the gut environment in summer months when the water is warmer (Korajkic et al., 2019). While Bacteroidetes are anaerobic and thus do not survive outside the host gut for long regardless of the season (Ahmed et al., 2014; Ballesté & Blanch, 2010; Kreader, 1998), E. coli can persist outside the host for longer periods in specific environments (Ishii & Sadowsky, 2008). Limitations Our findings serve to not only advance the understanding of water quality in coastal Connecticut but also help to highlight the limitations of using molecular markers to identify sources of fecal contamination. Major limitations include the lack of correlation between indicators and pathogens (Korajkic et al., 2018) and an inadequate understanding of the persistence of traditional and newer fecal indicators (Korajkic et al., 2019). We also acknowledge the limitations of the tested DNA sources. The tested bird sources did not pass quality assurance, which—along with the fact that other nontested bacteria sources (e.g., rodents, etc.) might have been present in the sample—means that at this time it is not possible to correlate E. coli with bacteria-specific sources of bacterial contamination. Further research into source tracking as a means of determining public health risk is warranted. An additional or alternate direction for future studies could be to employ next-generation sequencing technologies to assess likely sources of bacteria and to detect actual pathogens rather than focusing on surrogate indicators. Moreover, a series of sampling points along each river in conjunction with a more aggressive sampling schedule that included precipitation events would have been the preferred collection methodology. Due to limited project resources, however, a single sample location was selected for each targeted watershed, with collections conducted in such a way as to avoid tidal influence. Conclusion This study confirmed past findings that the targeted watersheds were consistently ašected by elevated levels of fecal contamination after a rainfall event as detected by the indicators E. coli and GenBac3. In addition, through the use of host-associated molecular markers used for microbial source tracking, we found no evidence to support the hypothesis that any of the sites were chronically impacted by human, ruminant (including cattle), poultry, or canine feces. Absence and Presence of Human-Associated (A) and RuminantAssociated (B) Markers in the Three Targeted Watersheds Note. The presence of the human-associated and ruminant-associated markers did not correlate with elevatedE. coli levels. Goodwives River Sasco Brook 1 2 3 4 1 2 3 4 Human Marker Jan 2016 Apr 2016 Jul 2016 Oct 2016 Jan 2016 Apr 2016 Jul 2016 Oct 2016 Jan 2016 Apr 2016 Jul 2016 Oct 2016 Jan 2016 Apr 2016 Jul 2016 Oct 2016 Jan 2016 Apr 2016 Jul 2016 Oct 2016 Jan 2016 Apr 2016 Jul 2016 Oct 2016 Absent Present Ruminant Marker Absent Present Lower Farm River, Branford B Log10 E. coli (CFU/100 ml) Log10 E. coli (CFU/100 ml) A Lower Farm River, Branford FIGURE 3